LECTURE #10 A. The Hardy-Weinberg Equilibrium 1. From the definitions of p and q, and of p 2, 2pq, and q 2, an equilibrium is indicated (p + q) 2 = p 2 + 2pq + q 2 : if p and q remain constant, and if all genotypes mate randomly with all other genotypes (as a function of their frequency in the population), the values of p 2, 2pq, and q 2 will remain constant generation after generation after generation a) Try mating: p 2 + 2pq + q 2 Note: there is random mating (panmixia) and p 2 + 2pq + q 2 no forces acting to change gene frequency b) This is the Hardy-Weinberg equilibrium and represents the expected frequencies of genotypes in every generation if and only if there are No forces & Random mating c) The equilibrium is the cornerstone of population genetics and the very first thing a population geneticist does after estimating allele and genotype frequencies is test whether the population is in the equilibrium at the locus (loci) under study 2. How does one test whether a population is in the HW equilibrium? a) Use a Chi-square goodness-of-fit test and determine whether observed numbers of each genotype fit expected numbers (based on equilibrium proportions) (i) try the following example: Genotype Observed AA 50 Aa 0 aa 50 3. If a population is not in HW equilibrium, how long will it take the population to return to the equilibrium? a) For autosomal loci, one generation (regardless of number of alleles) if the underlying assumptions of the equilibrium (i.e., no forces and random mating) are operative (i) try the above example, with random mating 4. Note the following: a) If deviations from the HW equilibrium are due to non-random mating, there will be changes in genotype frequency but not changes in gene (allele) frequencies; the changes are usually increases in frequencies of homozygous genotypes and decreases in heterozygous genotype b) If deviations from the HW equilibrium are due to a force, there will be changes in both gene (allele) and genotype frequencies
2 B. Calculation of allele frequencies when there are multiple alleles 1. Try example with ABO blood group system Use the following: p = I A, q = I B, and r = I 2. At equilibrium, (p + q + r) 2 = p 2 + 2pq + 2pr + q 2 + 2qr + r 2 a) Frequency of A blood group phenotype is p 2 + 2pr Frequency of B blood group phenotype is q 2 + 2qr Frequency of AB blood group phenotype is 2pq Frequency of O blood group phenotype is r 2 b) A sample from a population: Phenotype Number of individuals b) Frequency of Type O = r 2 = 360/1,000 r 2 = 0.36 r= 0.6 Type A 130 Type B 450 Type AB 60 Type O 360 1,000 c) Then, p 2 + 2pr + r 2 = frequency of type A + type O (p + r) 2 = 360 + 130/1,000 (p + r) 2 = 0.49 p + r = 0.7 p = 0.1 d) Can do the same for q, or deduce simply that q = 0.3 C. The problem with estimating allele frequencies for recessives 1. Necessary to assume population is in HW equilibrium a) Example: simple recessive, where incidence of homozygous recessives is 25% in a sample of 100 individuals Number of AA + Aa = 75 Number of aa = 25 b) Let q 2 = 25/100, q = 0.5, q = p c) Find method to score heterozygotes and find #AA = 65, #Aa = 10
3 d) Estimate q as: q = 25 (2) + 10 (1)/200 q = 60/200 q = 0.3 q = 60/200 q = 0.3, p = 0.7 e) At equilibrium, p 2 (AA) = 0.49 2pq (Aa) = 0.42 q 2 (aa) = 0.09 f) Note that the population was not in the HW equilibrium (as assumed), and moreover, that the allele frequencies were not p = q = 0.5, but rather p = 0.7 and q = 0.3 (and with different equilibrium genotype expectations) g) Note also that for deleterious recessives, selection is already implied (so how could one expect a population to be in HW equilibrium) Inbreeding and Assortative Mating: Deviations from Random Mating (Panmixia) A. Inbreeding 1. Non-random mating between individuals related by descent 2. A two-edged sword: main effects are to a) Increase homozygosity at the expense of heterozygosity, leading to inbreeding depression (generally bad) b) Remove (cull) deleterious recessive alleles (generally good) 3. Loss of heterozygosity and increase of homozygosity is best demonstrated by self-fertilization, the most intense form of inbreeding: consider a starting population where p = q = 0.5, and S 1, S 2, S n refer to successive generations AA (1/4) Aa (1/2) aa (1/4) S 1 p = q = 0.5 AA (3/8) Aa (2/8) aa (3/8) S 2 p = q = 0.5 AA (7/16) Aa (2/16) aa (7/16) S 3 p = q = 0.5 a) Heterozygosity decreased by1/2 each generation b) Gene frequencies remain constant c) Genotype frequencies change in characteristic pattern (homozygotes increase at expense of heterozygotes 4. The rate at which heterozygosity decreases (homozygosity increases) is strictly a function of the degree of inbreeding
4 5. Oddly, while decreasing genetic variability (heterozygosity), inbreeding can lead to increased phenotypic diversity: this occurs under situations where a population becomes subdivided a) Examples included domesticated canines and felines -- observe loss of genetic variability (heterozygosity) through inbreeding and significant increase of genetic abnormalities, yet see dramatic increase in phenotypic diversification through generation of multiple (even extreme) breeds B. Consanguinity 1. Consider pedigree of phenylketonuria (inherited as a simple recessive) involving a marriage between second cousins: Allele frequencies: p (pku + ) = 0.99 q (pku) = 0.01 2. Consider the following questions: a) Probability of affected child if IV-2 marries IV-3 (second cousin marriage)? b) Probability of affected child if IV-2 marries normal person at random? c) Probability of affected child if two normal people marry at random? For (a), probability is not related to the frequency of the pku allele in the population (1/6) (1/6) (1/4) = about 1% Note: about 3% for first cousin marriages For (b), probability is 1/6 (IV-2) and 2pq (0.0918) (1/6) (0.0198) (1/4) = about 0.08% For (c), probability is 2pq x 2pq (0.0198) 2 (1/4) = about 0.01% d) Ratios: (i)/(ii) = second cousin marriage verses one individual at random = 12x (i)/(iii) = second cousin marriage versus two individuals at random = 102X Note that the probabilities obtained for ratios (b) and (c) will decrease as the frequency of the q (pku) allele becomes smaller but that the probability obtained for ratio (a) remains the same (as it is not a fuction of the frequency of the recessive allele in the population.)
5 C. Assortative mating 1. Deviations from random mating where individuals select mates based on phenotypes or genotypes a) Phenotypic assortative mating positive b) Phenotypic assortative mating negative c) Genotypic assortative mating positive d) Genotypic assortative mating negative 2. Most known instances generally involve positive assortative mating (like genotypes, phenotypes): genetic impact in increase in homozygosity at expense of heterozygosity but only for loci that are directly involved in the assortative mating differs from inbreeding in that the latter is a genome wide effect